UAV Flight Research Articles

Application of Reinforcement Learning in Controlling Quadrotor UAV Flight Actions

Most literature has extensively discussed reinforcement learning (RL) for controlling rotorcraft drones during flight for traversal tasks. However, most studies lack adequate details regarding the design of reward and punishment mechanisms, and there is a limited exploration of the feasibility of applying reinforcement learning in actual flight control following simulation experiments. Consequently, this study focuses on the exploration of reward and punishment design and state input for RL. The simulation environment is constructed using AirSim and Unreal Engine, with onboard camera footage serving as the state input for reinforcement learning. The research investigates three RL algorithms suitable for discrete action training. The Deep Q Network (DQN), Advantage Actor–Critic (A2C), and Proximal Policy Optimization (PPO) were combined with three different reward and punishment design mechanisms for training and testing. The results indicate that employing the PPO algorithm along with a continuous return method as the reward mechanism allows for effective convergence during the training process, achieving a target traversal rate of 71% in the testing environment. Furthermore, this study proposes integrating the YOLOv7-tiny object detection (OD) system to assess the applicability of reinforcement learning in real-world settings. Unifying the state inputs of simulated and OD environments and replacing the original simulated image inputs with a maximum dual-target approach, the experimental simulation achieved a target traversal rate of 52% ultimately. In summary, this research formulates a set of logical frameworks for an RL reward and punishment design deployed with real-time Yolo’s OD implementation synergized as a useful aid for related RL studies.

Real-Time Planning of Minimum-Time Trajectories for Agile UAV Flight

Real-Time Planning of Minimum-Time Trajectories for Agile UAV Flight

Retraction Note: An agent architecture for autonomous UAV flight control in object classification and recognition missions

Retraction Note: An agent architecture for autonomous UAV flight control in object classification and recognition missions

Solving UAV 3D Path Planning Based on the Improved Lemur Optimizer Algorithm

This paper proposes an Improved Lemur Optimization algorithm (ILO), which combines the advantages of the Spider Monkey Optimization algorithm, Simulated Annealing algorithm, and Lemur Optimization algorithm. Through the use of an adaptive nonlinear decrement model, adaptive learning factors, and updated jump rates, the algorithm enhances its global exploration and local exploitation capabilities. A Gaussian function model is used to simulate the mountain environment, and a mathematical model for UAV flight is established based on constraints and objective functions. The fitness function is employed to determine the minimum cost for avoiding obstacles in a designated airspace, and cubic spline interpolation is used to smooth the flight path. The Improved Lemur Optimization algorithm was tested using the CEC2017 benchmark set, assessing its search capability, convergence speed, and accuracy. The simulation results show that ILO generates high-quality, smooth paths with fewer iterations, overcoming the issues of premature convergence and insufficient local search ability in traditional genetic algorithms. It adapts to complex terrain, providing an efficient and reliable solution.

Air Route Design of Multi-Rotor UAVs for Urban Air Mobility

UAVs will present significant air traffic in the urban airspace in the future, which brings new challenges to urban air traffic management and control. This paper presents an air route design scheme for multi-rotor UAVs in urban airspace to enable UAV operations at orderly levels. The air routes include legs and intersections, which are the three-dimensional channels of UAV flight. Based on the concept of structured and layered urban airspace, the cylindrical pipeline flight leg is designed, and the operation concept, characteristic parameters and flight procedures of along-road and roundabout intersections are proposed. By defining UAV conflict risk and intersection service level metrics, the operation situation of UAVs is quantitatively evaluated. Taking an urban transportation scenario as a case, the proposed route design scheme is simulated in different scale UAV operating scenarios. The results show that the number of UAVs at the intersection is positively correlated with the conflict probability, the number of crossing routes is negatively correlated with the intersection passing rate, and the UAV arrival rate is positively correlated with the intersection average passing time. The along-road type intersection is suitable for the area with fewer crossing routes and sparse UAVs, while the roundabout type intersection is adapted for the area with more crossing routes and dense UAVs. This research provides a new idea for urban UAV air route design, which is helpful in promoting the standardized management of UAVs and accelerating the integration of UAVs into urban airspace.

UAV Hunter: A Net-Capturing UAV System with Improved Detection and Tracking Methods for Anti-UAV Defense

The abuse of UAVs poses a potential risk to social security, necessitating the investigation of anti-UAV methods to safeguard critical areas. However, the existing UAV countermeasures face challenges such as high environmental impact, restricted spatial deployment, and low cost-effectiveness. To address these limitations, we developed a novel anti-UAV system known as UAV Hunter, which adopts an airborne tether-net capture device with visual aids to counter unauthorized UAVs. This system employs an “Anti-UAV with UAV” scheme, comprising a ground control station and a net-capturing UAV. The operator utilizes the ground control station to determine the mission area and flight path and then controls the flight of the net-capturing UAV. During flight, the net-capturing UAV leverages its dual-mode sensor to continuously monitor the target area. Simultaneously, the onboard computer executes a UAV detection and tracking algorithm to search for unauthorized UAVs in real time. The results are relayed to the operator in real time, facilitating precise adjustments for the net-capturing UAV to launch the rope net accurately. The system successfully realizes the functions of dual-mode real-time detection and tracking, precise net capture, and efficient integrated control. Compared with existing methods, the developed system exhibits accurate recognition, rapid action, diverse application scenarios, and an enhanced human–machine interaction experience. Test results in the open environment further validate the feasibility and functional integrity of the system, demonstrating its capability to effectively capture low-altitude unauthorized UAVs.

Recognition of maize seedling under weed disturbance using improved YOLOv5 algorithm

Using UAV-based RGB images to recognize maize seedlings is of great significant for precise weed control, efficient water and fertilizer management. However, the presence of weeds with morphological resemblances at the maize seedling stage affects the recognition of maize seedlings. This research employs UAV RGB images and deep learning algorithms to achieve accurate recognition of maize seedlings under weed disturbance. Firstly, the adaptive anchor frame algorithm is employed to intelligently select optimal anchor frame sizes suited for the maize seedling from UAV images. This strategic selection minimizes time and computational demands associated with multiple anchor frame sampling. Subsequently, the Global Attention Mechanism (GAM) is introduced, bolstering feature extraction capabilities. A range of deep learning models, including YOLOv3 and YOLOv5, are applied for maize seedling recognition, culminating in the identification of an optimal model. To account for real-world scenarios, we investigate the influences of UAV flight height and weed disturbance on maize seedling recognition. The results indicate a multi-class Average Precision (mAP) of 94.5 % and 88.2 % for detecting maize seedlings at flight heights of 15 m and 30 m, respectively, with an average detection speed of 0.025 s per single image. This emphasizes the efficacy of the improved YOLOv5 deep learning model in recognizing maize seedlings under weed disturbance using UAV RGB images.

Fixed-wing UAV obstacle avoidance algorithm based on real terrain

Abstract This paper presents an obstacle avoidance algorithm for UAVs based on real terrain constraints. The common obstacle avoidance algorithm often abstracts the obstacle as a sphere or a circle or studies the UAV’s two-dimensional obstacle avoidance on a raster map. With the development of fixed-wing UAVs, UAVs have longer and longer time in the air, longer and longer range, and the low-altitude environment they face is often difficult to simulate with simplified regular shape. Thus, this paper constructs the UAV flight terrain environment using open elevation data and designs an obstacle avoidance algorithm accordingly. When the UAV is tracking the track point, it can better control the UAV to avoid obstacles and fly to the target track point.

Investigation of airflow thermal ice melting process under various flight conditions

The challenge of preventing and removing ice from exposed regions of aircraft is a significant engineering concern. Understanding the melting process of ice due to aerodynamic heating during flight is essential for developing UAV flight strategies in icy conditions. This paper analyzes the heat transfer mechanisms involved in airflow over ice and introduces a theoretical model to estimate the ice melting rate, using principles of turbulent heat transfer at the stagnation point. The study also discusses the impact of environmental conditions on ice melting rates. Findings indicate that the melting speed of the ice surface exhibits a negative linear correlation with the initial temperature of the ice, whereas it shows a nonlinear correlation with airflow velocity and total temperature of the incoming flow. Lower airflow velocity or total temperature of the incoming flow enhances the sensitivity of ice melting speed to changes. Additionally, lower ice density results in a higher melting speed, showing an exponential relationship with factors like average droplet diameter, airflow velocity, and airfoil leading edge diameter in the cloud field during icing. Experiments conducted using a small jet test bench and an icing wind tunnel confirmed the impact of varying conditions on ice melting rates. The deviation between experimental results and theoretical predictions was under 5 %. These conclusions offer valuable insights for flight safety planning of unprotected iced aircraft in challenging environments.

3D Positioning of Drones through Images.

Drones traditionally rely on satellite signals for positioning and altitude. However, when in a special denial environment, satellite communication is interrupted, and the traditional positioning and height determination methods face challenges. We made a dataset at the height of 80-200 m and proposed a multi-scale input network. The positioning index RDS achieved 76.3 points, and the positioning accuracy within 20 m was 81.7%. This paper proposes a method to judge the height by image alone, without the support of other sensor data. One height judgment can be made per single image. Based on the UAV image-satellite image matching positioning technology, by calculating the actual area represented by the UAV image in real space, combined with the fixed parameters of the optical camera, the actual height of the UAV flight is calculated, which is 80-200 m, and the relative error rate of height is 18.1%.

Application of fuzzy logic control theory combined with target tracking algorithm in unmanned aerial vehicle target tracking

This paper aims to increase the Unmanned Aerial Vehicle's (UAV) capacity for target tracking. First, a control model based on fuzzy logic is created, which modifies the UAV's flight attitude in response to the target's motion status and changes in the surrounding environment. Then, an edge computing-based target tracking framework is created. By deploying edge devices around the UAV, the calculation of target recognition and position prediction is transferred from the central processing unit to the edge nodes. Finally, the latest Vision Transformer model is adopted for target recognition, the image is divided into uniform blocks, and then the attention mechanism is used to capture the relationship between different blocks to realize real-time image analysis. To anticipate the position, the particle filter algorithm is used with historical data and sensor inputs to produce a high-precision estimate of the target position. The experimental results in different scenes show that the average target capture time of the algorithm based on fuzzy logic control is shortened by 20% compared with the traditional proportional-integral-derivative (PID) method, from 5.2 s of the traditional PID to 4.2 s. The average tracking error is reduced by 15%, from 0.8 m of traditional PID to 0.68 m. Meanwhile, in the case of environmental change and target motion change, this algorithm shows better robustness, and the fluctuation range of tracking error is only half of that of traditional PID. This shows that the fuzzy logic control theory is successfully applied to the UAV target tracking field, which proves the effectiveness of this method in improving the target tracking performance.

Analysis of the impact of electromagnetic fields on UAV flight control systems in EHV–UHV DC overhead transmission lines

Analysis of the impact of electromagnetic fields on UAV flight control systems in EHV–UHV DC overhead transmission lines

Analysis of test results of the P-20 multi-rotor UAV when spraying pesticides in precision agriculture

Abstract The article provides an overview of the current results of using multi-rotor UAVs when spraying pesticides in precision agriculture. It is noted that, as part of improving the technology for using UAVs, it is necessary to continue work on optimizing and testing the parameters of UAV transport and technological cycles together with a reasonable choice of special reagents and their combinations to improve the spraying effect and increase the level of use of pesticides. This will directly affect the improvement of the quality and efficiency of precision farming systems. The analysis presented in the work is based on testing the parameters of the transport and technological cycle of a multi-rotor UAV of the P-20 type, which was used to spray pesticides on cotton fields to combat aphids and spider mites on cotton at the flowering and boll setting stages. The results are considered that make it possible to analyze and compare the laws of uniform and penetrating deposition and drift of droplets in cotton canopies at different UAV flight altitudes. It has been shown that spraying small volumes of agrochemicals using UAVs at low altitudes differs from operations performed using modern manned aircraft or boom sprayers. The flight altitude of the UAV has a significant impact on the deposition and drift of droplets. The results of studies related to elucidating the effect of defoliant adsorption on cotton leaves are also noted. The adsorption of acetamiprid and spirodiclofen was studied. Methods for statistical processing of the obtained experimental data are presented, in particular, to test the significance of the influence of flight altitude on the experimental results.

Optimization of UAV Flight Paths in Multi-UAV Networks for Efficient Data Collection

Optimization of UAV Flight Paths in Multi-UAV Networks for Efficient Data Collection

Analysis and evaluation of development characteristics of debris flow in Longshougou

Long Shougou is located at the exit of Hei he River in Zhang ye City, which is directly facing the reservoir of Long Shougou Hydropower Station. Through site survey, UAV flight and indoor simulation calculation, the development characteristics and risk of Long Shougou debris flow were analyzed and evaluated. The results show that: Long Shougou debris flow is in the "development period", belonging to "small", "easy" and "high frequency" debris flow; The length of Long Shougou Channel is 1.70 km, the circumference is 3.76 km, the area is 0.72 km2 , the relative height difference is nearly 584 m, the longitudinal gradient is 342.27 ‰, and the overall bending coefficient is 1.14; Gully sediments are mainly gully bed deposits and slope collapse deposits, with total reserves of 20.94 ×104 m3 and dynamic reserves of 7.76 ×104 m3 ; Calculated according to the designed rainstorm frequency of 1%, the flow velocity of debris flow at the gully mouth is 2.64m /s, the total amount of primary overflow is 17,296.41m3 , the primary outflow of solid matter is 7956.35m3 , and the overall impact pressure of debris flow is 16.32kPa. The maximum impact force of a single stone is 2.56kN according to 0.5m; It is suggested to consider the impact of the annual inflow of debris flow on the water quality of the Longshou hydropower station reservoir and the impact on the power station switching station near the Gulou port, and do a good job in the relevant sediment discharge measures.

Design of UAV Flight State Recognition System for Multisensor Data Fusion

Design of UAV Flight State Recognition System for Multisensor Data Fusion

UAV flight path planning optimization

UAV flight path planning optimization

Path Planning of Unmanned Aerial Vehicles Based on an Improved Bio-Inspired Tuna Swarm Optimization Algorithm.

The Sine-Levy tuna swarm optimization (SLTSO) algorithm is a novel method based on the sine strategy and Levy flight guidance. It is presented as a solution to the shortcomings of the tuna swarm optimization (TSO) algorithm, which include its tendency to reach local optima and limited capacity to search worldwide. This algorithm updates locations using the Levy flight technique and greedy approach and generates initial solutions using an elite reverse learning process. Additionally, it offers an individual location optimization method called golden sine, which enhances the algorithm's capacity to explore widely and steer clear of local optima. To plan UAV flight paths safely and effectively in complex obstacle environments, the SLTSO algorithm considers constraints such as geographic and airspace obstacles, along with performance metrics like flight environment, flight space, flight distance, angle, altitude, and threat levels. The effectiveness of the algorithm is verified by simulation and the creation of a path planning model. Experimental results show that the SLTSO algorithm displays faster convergence rates, better optimization precision, shorter and smoother paths, and concomitant reduction in energy usage. A drone can now map its route far more effectively thanks to these improvements. Consequently, the proposed SLTSO algorithm demonstrates both efficacy and superiority in UAV route planning applications.

Comparing Turbulence Models for CFD Simulation of UAV Flight in a Wind Tunnel Experiments

Wind tunnel tests are costly and time-consuming, and the accuracy of the tests is limited by the size of the tunnel, to solve this problem, researchers use computational fluid dynamics (CFD) to conduct wind tunnel experiments using a computer because it is less timely and less costly. Computerized testing of drone's models using wind tunnel experiment simulation in computational fluid dynamic (CFD) software requires knowledge of the most suitable turbulence model for this drone. In this paper ANSYS fluent program has been used to test four most common turbulence models for use (Spalart-Allmaras, K-Epsilon, K-Omega and Reynolds stress) and laminar flow on the ScanEagle drone model (an aerial reconnaissance drone used in military and intelligence operations) and calculated their effect on aerodynamic parameters In terms of accuracy and time to solution, concluded that the best turbulence model in terms of balancing accuracy and the time taken for the calculation is K-Omega model.

An Anomaly Detection Method for UAV Based on Wavelet Decomposition and Stacked Denoising Autoencoder

The paper proposes an anomaly detection method for UAVs based on wavelet decomposition and stacked denoising autoencoder. This method takes the negative impact of noisy data and the feature extraction capabilities of deep learning models into account. It aims to improve the accuracy of the proposed anomaly detection method with wavelet decomposition and stacked denoising autoencoder methods. Anomaly detection based on UAV flight data is an important method of UAV condition monitoring and potential abnormal state mining, which is an important means to reduce the risk of UAV flight accidents. However, the diversity of UAV mission scenarios leads to a complex and harsh environment, so the acquired data are affected by noise, which brings challenges to accurate anomaly detection based on UAV data. Firstly, we use wavelet decomposition to denoise the original data; then, we used the stacked denoising autoencoder to achieve feature extraction. Finally, the softmax classifier is used to realize the anomaly detection of UAV. The experimental results demonstrate that the proposed method still has good performance in the case of noisy data. Specifically, the Accuracy reaches 97.53%, the Precision is 97.50%, the Recall is 91.81%, and the F1-score is 94.57%. Furthermore, the proposed method outperforms the four comparison models with more outstanding performance. Therefore, it has significant potential in reducing UAV flight accidents and enhancing operational safety.